Gene silencing by RNA interference (RNAi) is rapidly becoming the method of choice for the elucidation of gene function and the identification of therapeutic targets. RNAi is envisioned to ultimately be used as a potent and selective human therapeutic. Short interfering RNAs (siRNA) can be chemically synthesized and do not require nuclear entry for efficacy. Delivery of siRNAs could potentially provide transient gene inhibition in humans for therapeutic use. However, effective systemic delivery of siRNA remains a challenge. Here, we investigate a new concept for the effective delivery of siRNA. siRNA duplexes containing phosphorothioate (PS) sense strands and phosphodiester (PO) antisense strands are formulated with proteins that are used to target cell surface receptors that provide endocytosis. Thus, the PS sense strand is employed to carry the antisense strand, impart stability to the duplex and to bind to proteins that can provide for cellular targeting and/or cellular uptake. No synthetic delivery systems are necessary for in vivo delivery. Hypothesis: A PS sense strand when combined with a PO antisense strand to form a siRNA duplex can elicit effective in vivo delivery of the duplex by binding to proteins that can provide cellular targeting and/or cellular uptake. Specific Aims: (1) Prepare siRNA with PS sense strands and PO antisense strands that give luciferase (LUC), FAS or EWS-FLI1 gene inhibition and investigate their in vitro behavior and their binding to proteins that can be ligands for cell surface receptors, e.g., asialofetuin (AF) for the asialoglycoprotein receptor on hepatocytes and transferrin (Tf) for transferrin receptor on cancer cells and the blood-brain barrier; (2) Measure the in vivo biocompatibility of the siRNA duplexes in mice by monitoring for toxicities and changes in blood chemistries; and (3) Measure the effectiveness of the siRNA for LUC, FAS and EWS-FLI1 gene inhibition by combining them with targeting proteins such as AF and Tf and using several dosing regimes in two mice models; namely, a transgenic model with liver luciferase expression and a metastatic tumor model that creates brain metastases to monitor the possibility of crossing of the blood-brain barrier. Phase II research will capitalize upon the findings in this Phase I program that is designed to define how to best target the PS(sense):PO(antisense) siRNA duplexes to liver and metastatic tumors and to define whether or not they cross the blood-brain barrier. Based on the results from the Phase I study, disease targets in the locations successfully reached will be explored. For example, we know from our preliminary results that hepatocytes can be targeted with this technology. Disease targets in the liver will be investigated in Phase II as well as possible tumor and brain targets. The expected outcome of the Phase II investigation will be a siRNA that can be taken into a human clinical study. [unreadable] [unreadable] [unreadable]